Dependences of photoluminescence from P-implanted epitaxial Ge

doi:10.1364/OE.20.008228

« Show journal navigation

Dependences of photoluminescence from P-implanted epitaxial Ge

L. Ding, Andy Eu-Jin Lim, Jason Tsung-Yang Liow, M. B. Yu, and G.-Q. Lo »View Author Affiliations

Optics Express, Vol. 20, Issue 8, pp. 8228-8239 (2012)
http://dx.doi.org/10.1364/OE.20.008228

View Full Text Article

Acrobat PDF (1175 KB)

Journal Search
Article Lookup

Article Tools

Citations

Alert me when this article is cited
Export Citation/Save

Article Navigation

▸ Article Outline
– Abstract
1. Introduction
2. Experiment
3. Results and Discussion
4. Summary
– References and links

Article Tools
Full Text
Article Info
References
Cited By
Figures
Metrics
Download PDF

Viewing Equations in the
HTML Article

Optimal Viewing Recommendations

Full Text
Article Info
References (22)
Cited By
Figures (11)
Metrics

Abstract

A systematic investigation has been carried out to study the influence of various annealings and implantations on the photoluminescence (PL) properties of phosphorus (P)-implanted Ge epitaxial films on Si substrate. For un-capped Ge samples, rapid thermal annealing (RTA) at 700 °C for 300 seconds yields the strongest PL emission peaked at 1550 nm. The influence of employing various capping layers (i.e., SiO₂, Si₃N₄, and α-Si) on the PL properties has been investigated. The capping layers are found to effectively decrease the dopant loss, leading to a significant PL enhancement. Si₃N₄ is found to be the most efficient capping layer to prevent dopant out-diffusion and thus lead to strongest PL. Furthermore, it has been found that capping layers not only enhance the PL intensities but also make PL emission peak red- and blue- shift, depending on the stress type of the capping films. The effect of implantation dose on the PL has been also investigated.

1. Introduction

A light-emitting device (LED) that can be monolithically integrated with complementary metal-oxide-semiconductor (CMOS) circuit and other Si photonic devices is a missing component to achieve all-Si CMOS photonics. Germanium (Ge) has been investigated intensively as a Si-compatible photodetector (PD) material [1

L. Chen, P. Dong, and M. Lipson, “High performance germanium photodetectors integrated on submicron silicon waveguides by low temperature wafer bonding,” Opt. Express 16(15), 11513–11518 (2008). [CrossRef] [PubMed]

–3

K.-W. Ang, M.-B. Yu, G.-Q. Lo, and D.-L. Kwong, “Low voltage and high responsivity germanium bipolar phototransistor for optical detections in the near-infrared regime,” IEEE Electron Device Lett. 29(10), 1124–1127 (2008). [CrossRef]

]. Recently, Ge has also been theoretically predicted to be a promising material to achieve ~1.55 µm light emission leveraging CMOS process so that the Si chip can be directly connected to optical fiber network [4

J. Liu, X. Sun, D. Pan, X. Wang, L. C. Kimerling, T. L. Koch, and J. Michel, “Tensile-strained, n-type Ge as a gain medium for monolithic laser integration on Si,” Opt. Express 15(18), 11272–11277 (2007). [CrossRef] [PubMed]

]. Since then, intensive effort has been devoted into the research of Ge direct bandgap light emission, and some experimental results including both photoluminescence (PL) [5

X. Sun, J. Liu, L. C. Kimerling, J. Michel, and T. L. Koch, “Direct gap photoluminescence of n-type tensile-strained Ge-on-Si,” Appl. Phys. Lett. 95(1), 011911 (2009). [CrossRef]

–8

T.-H. Cheng, K.-L. Peng, C.-Y. Ko, C.-Y. Chen, H.-S. Lan, Y.-R. Wu, C. W. Liu, and H.-H. Tseng, “Strain-enhanced photoluminescence from Ge direct transition,” Appl. Phys. Lett. 96(21), 211108 (2010). [CrossRef]

] and electroluminescence (EL) [9

X. Sun, J. Liu, L. C. Kimerling, and J. Michel, “Room-temperature direct bandgap electroluminesence from Ge-on-Si light-emitting diodes,” Opt. Lett. 34(8), 1198–1200 (2009). [CrossRef] [PubMed]

,10

S.-L. Cheng, J. Lu, G. Shambat, H.-Y. Yu, K. Saraswat, J. Vuckovic, and Y. Nishi, “Room temperature 1.6 microm electroluminescence from Ge light emitting diode on Si substrate,” Opt. Express 17(12), 10019–10024 (2009). [CrossRef] [PubMed]

] have been reported. Although Ge is an indirect bandgap semiconductor, the energy difference between its direct bandgap (i.e., 0.8 eV) and indirect bandgap (i.e., 0.664 eV) is only 0.136 eV, which makes it achieve a great success in PD applications. It is interesting to note that the direct bandgap of Ge is 0.8 eV, corresponding exactly to 1550 nm which is commercially used in fiber optical communication. From previous reported studies on direct bandgap light emission from Ge, it is found that two important properties, i.e., active n-type doping level and in-plane tensile strain, would have significant effect on the light emission efficiency of single crystalline Ge film [4

–10

]. By introducing n-type dopant and in-plane tensile strain in Ge, the energy difference between its direct bandgap at Γ valley and indirect bandgap at L valley becomes smaller, and thus the Ge becomes closer to a direct bandgap material [4

–6

J. Liu, X. Sun, L. C. Kimerling, and J. Michel, “Towards a Ge-based laser for CMOS applications,” in Proceedings of 5^th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, Italy, 2008), pp. 16–18.

]. Due to the less lattice damage introduced, in situ doping method has been chosen to fabricate n-type Ge films by a few pioneer research groups working on light-emitting Ge [5

X. Sun, J. Liu, L. C. Kimerling, J. Michel, and T. L. Koch, “Direct gap photoluminescence of n-type tensile-strained Ge-on-Si,” Appl. Phys. Lett. 95(1), 011911 (2009). [CrossRef]

–7

M. El Kurdi, T. Kociniewski, T.-P. Ngo, J. Boulmer, D. Débarre, P. Boucaud, J. F. Damlencourt, O. Kermarrec, and D. Bensahel, “Enhanced photoluminescence of heavily n-doped germanium,” Appl. Phys. Lett. 94(19), 191107 (2009). [CrossRef]

,10

]. In CMOS industry, however, implantation of phosphorus (P) or arsenic (As) followed by rapid thermal annealing (RTA) is commonly used to introduce n-type dopant [11

C. O. Chui, K. Gopalakrishnan, P. B. Griffin, J. D. Plummer, and K. C. Saraswat, “Activation and diffusion studies of ion-implanted p and n dopants in germanium,” Appl. Phys. Lett. 83(16), 3275–3277 (2003). [CrossRef]

–14

A. Chroneos, D. Skarlatos, C. Tsamis, A. Christofi, D. S. McPhail, and R. Hung, “Implantation and diffusion of phosphorous in germanium,” Mater. Sci. Semicond. Process. 9(4-5), 640–643 (2006). [CrossRef]

]. Therefore, it is necessary to do optimization of annealing conditions to achieve best luminescence performance of P-implanted Ge. Such an effort has not been reported so far.

In this study, high quality epitaxial Ge thin films were grown on Si wafers by ultrahigh vacuum chemical vapor deposition (UHVCVD) followed by P implantation. Various annealings, i.e., various annealing temperatures and annealing durations, with and without capping layers, have been employed to re-crystallize the P-implanted Ge and activate the n-type dopant. Sheet resistance was measured by four-point-probe to monitor the dopant activation in Ge. Annealing at 700 °C for 300 seconds was found to be the optimized annealing condition which yields the strongest PL intensity at 1550 nm. Moreover, the emission intensity was further enhanced by depositing a capping layer onto the P-implanted Ge before annealing. Meanwhile, the PL spectra were found to be red- or blue- shifted due to the presence of capping layer during annealing, and the shifts were depend on the stress of the capping film. Based on above findings, an optimized process flow is proposed to induce strongest PL from P-implanted Ge.

2. Experiment

Substrates that were used for Ge epitaxy were 8-inch p-type Si (100) wafers. The epitaxial growth of Ge was performed in an ultra-high vacuum chemical vapor deposition (UHVCVD) reactor commenced with the expitaxy of a 20-nm SiGe buffer and a 30-nm pure Ge seed layer at 400 °C. Afterwards, ~200-nm pure Ge epitaxy was carried at 550 °C. Finally, a 5-nm-Si layer was grown on top of Ge for passivation. The detailed Ge epilayer growth was described and studied elsewhere [15

T. H. Loh, H. S. Nguyen, C. H. Tung, A. D. Trigg, G. Q. Lo, N. Balasubramanian, D. L. Kwong, and S. Tripathy, “Ultrathin low temperature SiGe buffer for the growth of high quality Ge epilayer on Si(100) by ultrahigh vacuum chemical vapor deposition,” Appl. Phys. Lett. 90(9), 092108 (2007). [CrossRef]

]. Two-energy (i.e., 80 keV and 30 keV) P-implantations were carried out with various doses to introduce n-type dopant in the Ge epi-film. Implantations at two different energies were employed to induce relatively uniform distribution of P in the implanted area of Ge. Before the dopant activation annealing, some of the samples are deposited with different capping films, i.e., SiO₂, Si₃N₄, and α-Si, as dopant diffusion blocking layers. The stresses of the passivating films were measured, i.e., 21-MPa compressive stress for plasma enhanced chemical CVD (PECVD) deposited SiO₂, 310-MPa tensile stress for PECVD deposited Si₃N₄, and 360-MPa for low pressure CVD (LPCVD) deposited α-Si. Post-implantation annealings were carried out in N₂ ambient at temperatures of 500-800 °C for durations of 30-900 sec to activate dopants and recover the lattice damage caused by implantation.

The PL measurements were carried out at room temperature using a Renishaw micro-PL system. The excitation source is a diode laser emitting at 785 nm with incident power of 1 mW. The light emissions spectra were collected from wavelength of 1200 to 1600 nm by a GaAs detector cooled in liquid N₂ in the vertical direction through the microscope.

3. Results and Discussion

3.1. Annealing effect on material revolution

The as-grown Ge film presents high surface quality as measured by Atomic Force Microscopy (AFM), showing that the root mean square (RMS) surface roughness is less than 0.5 nm within a scanned area of 5×5 µm², as can be seen in Fig. 1(a) . Figure 1(b) shows the transmission electron microscopy (TEM) of as-grown Ge film, and the inset shows the electron diffraction pattern, indicating the single crystallinity nature of the as-grown film. After implantation, two phase of Ge was observed in the film. For example, Fig. 1(c) shows the cross-section TEM image of the as-implanted Ge film with a total dose of 2×10¹⁵ cm⁻². An amorphous Ge layer can be observed after implantation due to the lattice damage caused by high energy implanted ions. The Ge film beyond the ion projection range shows a crystalline phase since it was not attacked by the implanted ions, as shown in the right of Fig. 1(c). Thermal annealing was performed to activate dopants as well as recover the crystalline phase of Ge and reduce the thread dislocation defects. For example, Fig. 1(d) shows the TEM image of the P-implanted Ge caped with 100-nm-SiO₂ after 700-°C annealing for 300 seconds. It can be seen that the amorphous phase of Ge shown in Fig. 1(c) disappears and crystalline phase is present in the whole film.

Fig. 1 (a) Surface AFM image of as-grown Ge film on Si; (b) cross-sectional TEM image of as-grown Ge film; (c) cross-sectional TEM image of as-implanted Ge film; (d) cross-sectional TEM image of P-implanted Ge film annealing at 700 °C for 5 min with 100-nm-SiO₂ capping layer.

The as-implanted samples were thereafter annealed at different temperatures for various durations in N₂ ambient without a capping layer. Micro-Raman spectroscopy was performed with a 488-nm laser. Figure 2 shows the Raman spectra of the as-grown, as-implanted, and annealed samples at different temperatures for a fixed duration of 5 min. For as-grown Ge samples, Si-Si optical photon mode can be observed due to the presence of 5-nm Si capping layer on top of Ge for the Ge surface passivation. It can be seen than α-Ge phase dominates the Raman spectrum for the as-implanted sample, being consistent with the TEM image shown in Fig. 1(a). After annealing, the Raman spectra are dominated by Ge–Ge optical phonon mode near 300 cm⁻¹, due to the recovery of crystalline phase by annealing. Interestingly, as shown in the figure, Raman spectra present that there is Si-Ge photon mode enhanced by increasing thermal annealing. This is because of the intermixing of Si cap layer with Ge film during thermal heating. As shown in the figure, the Ge-Ge phonon intensity, which is the indicator of crystalline Ge content, increases as the annealing temperature goes higher. The Ge-Ge phonon intensity is even stronger than the as-grown sample when the P⁺-implanted Ge experiences an annealing at 700 °C or higher than 700 °C, indicating the annealing at 700 °C or above is adequate to make a phase transition from amorphous state to crystalline state for P⁺-implanted Ge epi-film. When the annealing temperature is below 700 °C, the Raman spectra show a large contribution from α-Ge, indicating that annealing below 700 °C cannot induce a fully recovery to crystalline state.

Fig. 2 Raman spectra of as-grown, as-implanted, and annealed P-implanted Ge epitaxial films on Si. The samples were annealed from 500 °C to 800 °C for a fixed duration of 5 min without a capping layer.

There have been studies showing that the tensile strain in epitaxial Ge on Si is enhanced with higher annealing temperature [4

,16

G. Grzybowski, R. Roucka, J. Mathews, L. Jiang, R. T. Beeler, J. Kouvetakis, and J. Menendez, “Direct versus indirect optical recombination in Ge films grown on Si substrate,” Phys. Rev. B 84(20), 205307 (2011). [CrossRef]

,17

D. D. Cannon, J. Liu, Y. Ishikawa, K. Wada, D. T. Danielson, S. Jongthammanurak, J. Michel, and L. C. Kimerling, “Tensile strained epitaxial Ge films on Si(100) substrates with potential application in L-band telecommunications,” Appl. Phys. Lett. 84(6), 906–908 (2004). [CrossRef]

]. As well known, this is attributed to the coefficient of thermal expansion (CTE) mismatch between Si and Ge. In this study, Raman analysis has been performed to verify the strain evolution with annealing. Figure 3(a) shows the high resolution Raman spectra of P-implanted Ge epi-films with various annealing temperatures, and the in-plane tensile strain is calculated from the strain versus phonon peak shift relationship Δω = be_xx, where the proportionality factor b = −408 cm⁻¹ and the phonon deformation potentials and elastic stiffness coefficients are from Refs [18

F. Cerdeira, C. Buchenauer, F. Pollak, and M. Cardona, “Stress-induced shifts of first-order Raman frequencies of diamond- and Zinc-blende-type semiconductors,” Phys. Rev. B 5(2), 580–593 (1972). [CrossRef]

,19

S. S. Mitra and N. E. Massa, Handbook of Semiconductors, T. S. Moss ed. (North-Holland, 1986) Vol. 1, p. 96.

]. It is found that the tensile strain Ge changes not as significantly as in Ref [17

], i.e., from 0.11% at 500 °C to 0.13% at 800 °C, as shown in Fig. 3(b). For comparison, the strain evolution with thermal annealing extracted from Ref [17

]. is also included in the figure. There are three major differences in the sample fabrication between our study and Refs [4

,16

,17

], which would contribute to the different observations in the strain evolution with annealing. Firstly, to improve the Ge quality, our method included a graded SiGe buffer between Si and Ge which significantly reduces the residue strain in the grown Ge. As can be seen in Fig. 2, the SiGe signal becomes stronger with thermal annealing. This buffer layer reduces the effect of thermal annealing on the tensile strain. Secondly, implantation was used to introduce n-type dopant in this study, while the above three studies used in situ doping method. Implantation first caused the Ge epitaxial layer to be amorphous, and subsequent annealing process re-crystallized the film and activated the dopant while dopant loss is also happening at the same time. It is a very complicated process involving material phase transition, defect annihilation and creation, dopant activation and loss, and stress evolution, all of which affect or interact with others. The strain evolution with annealing is much more complicated than in the Ge film with in situ dopant. This complexity would hide the mechanism that contributes to the insignificant strain evolution with thermal annealing. Thirdly, the Ge layer (230 nm) in this study is much thinner than that in Ref [4

,16

,17

], i.e., over 1 µm. There has been a strong argument that the tensile strain is larger in the thicker Ge film than thinner Ge film grown on Si, though it is not fully understood from point view of growth kinetics [20

J. Joo, S. Kim, I. G. Kim, K. S. Jang, and G. Kim, “High-sensitivity 10 Gbps Ge-on-Si photoreceiver operating at lambda approximately 1.55 microm,” Opt. Express 18(16), 16474–16479 (2010). [CrossRef] [PubMed]

]. The very small Ge thickness would be another reason that the strain evolution is insignificant with thermal annealing.

Fig. 3 (a) high resolution Raman spectra of P-implanted Ge epi-film with various annealings; (b) in-plane tensile strain in the epitaxial Ge film as the function of annealing temperature.

Since the active electron density in the film influences the film resistivity, the sheet resistance is an important indicator of the dopant activation in Ge epi-film. It is known that lower sheet resistance corresponds to higher electron density in n-type semiconductor, thus meaning more efficient dopant activation process occurs. In this study, the sheet resistance was measured by four-point probe to monitor the dopant activation with thermal annealing. Figure 4(a) presents the sheet resistance of the film as a function of annealing temperature for the P⁺-implanted Ge epi-film, and the sheet resistance of as-grown and as-implanted sample are also included. The large increase in sheet resistance from as-grown to as-implanted film, as shown in the figure, is because of the formation of amorphous Ge phase due to the implantation, which was confirmed by TEM and Raman spectroscopy. As shown in Fig. 4(a), the sheet resistance is decreased dramatically after annealing at 500 °C due to the dopant activation and partial recovery of crystal structure which was confirmed by Raman spectroscopy. The further decease in sheet resistance from annealing temperature of 500 to 700 °C is due the dopant activation increasing with annealing temperature [11

] as well as the fully recovery of α-Ge state to crystalline state, leading to the increase of electron density. The sheet resistance increases while the annealing temperature is above 700 °C. This is due to the reduction in active dopant density in result of increased dopant loss at higher annealing temperatures, which has been reported in a few studies [11

–14

]. Considering the dopant activation efficiency and Ge crystal recovery, 700 °C is therefore regarded as the optimized annealing temperature for making P⁺-implanted Ge to emit most efficiently. The evolution of sheet resistance has been also monitored with annealing time at the fixed temperature of 700 °C. It has been found that 300-sec annealing induces the lowest sheet resistance, as shown in Fig. 4(b). The increase in sheet resistance with the increasing annealing time from 300 sec is because that much more severe dopant loss happens although the Ge crystalline quality may improve. Therefore, one can believe that 700-°C annealing for 5 min induces the highest electron density in Ge. The effective electron density corresponds to sheet resistance, thus the value can be roughly estimated according to the chart given in Ref [21

S. M. Sze and J. C. Irvin, “Resistivity, mobility and impurity levels in GaAs, Ge, and Si at 300 °K,” Solid-State Electron. 11(6), 599–602 (1968). [CrossRef]

], assuming that the excess electrons are distributed uniformly in Ge. The average effective electron density is estimated to be 2 × 10¹⁹ cm⁻³ after 700-°C annealing for 300 seconds. As a total dose of 2 × 10¹⁵ cm⁻² of P ions has been implanted into 230-nm Ge, the activated P is only ~23% of the total implanted ions, and thus the total dopant loss including the unactivated and out-diffused ions is estimated to be ~77%.

Fig. 4 Sheet resistance of P⁺-implanted Ge epi-film without a capping layer as functions of (a) annealing temperature for a fixed duration of 300 sec; and (b) annealing time at a fixed temperature of 700 °C.

3.2. Annealing effect on the photoluminescence

For the un-capped samples, Fig. 5(a) shows the PL spectra of the samples annealed at 500-800 °C for a fixed duration of 300 seconds, and Fig. 5(b) shows the PL spectra of the samples annealed for 30 to 900 seconds at a fixed temperature of 700 °C. The result show broad emission bands peaked at 1550 nm in consistence with the Ge direct band gap, i.e., 0.8 eV. The peak wavelength shown in this work is shorter than the value of PL reported earlier by Liu et.al (i.e., 1600 nm) using P-implantation and in situ doping method [5

X. Sun, J. Liu, L. C. Kimerling, J. Michel, and T. L. Koch, “Direct gap photoluminescence of n-type tensile-strained Ge-on-Si,” Appl. Phys. Lett. 95(1), 011911 (2009). [CrossRef]

]. There are two possible reasons that probably contribute to the shorter emission peaks as compared with Liu’s work. The first reason is that our Ge epitaxy technique includes a SiGe buffer layer before Ge growth which causes stress relaxation leading to a layer of subsequent strain-free Ge film. The second contributor would be originated from the bandgap expansion due to the SiGe formed on the Ge surface by the intermixing effect of Ge and Si during annealing as there is a 5-nm Si capping layer growth for Ge passivation immediately after Ge epitaxy. The formation of SiGe alloy can be confirmed by Raman spectra, as shown in Fig. 2. Raman spectra present that there is Si-Ge photon mode enhanced by increasing thermal annealing. This is because of the intermixing of Si capping layer with Ge film during thermal heating, which would lead to a small blueshift of Ge direct band gap emission. As can be seen in Figs. 5(a) and 5(b), the strongest PL emission is observed from the sample annealed at 700-°C for 300 seconds. The lowest sheet resistance (i.e., highest electron density) of the sample annealed at 700 °C for 300 seconds as shown in Fig. 3 would explain the strongest PL intensity occurs at the annealing temperature of 700 °C for the duration of 300 seconds. For a clearer demonstration, the insets of Figs. 5(a) and 5(b) present the revolution of PL peak intensity with the annealing temperature and time, respectively. In Ref [16

], it was presented that the Ge direct bandgap PL peak would have red-shift due to the increased tensile strain caused by thermal annealing at 725 °C. However, the PL peak position presents no obvious shift with thermal annealing as shown in Fig. 5. As revealed in Fig. 3(b), with thermal annealing the strain of the Ge in our study increases much less than the Ge grown without SiGe buffer layer. The much smaller strain in our study would explain the unobservable PL peak shift. The other important reason is that the intermixing of Si capping layer and Ge epi-film becomes more severe with enhanced annealing process.

Fig. 5 (a) PL spectra for the P⁺-implanted Ge epi-films annealed at various temperatures for a fixed duration of 300 seconds. The inset shows the PL peak intensity as a function of annealing temperature. (b) PL spectra for the P⁺-implanted Ge epi-films annealed for various durations at a fixed temperature of 700 °C. The inset shows the PL peak intensity as a function of annealing time.

3.3. Effect of capping layers on the photoluminescence

Although strongest PL emission has been obtained with the optimized annealing, significant dopant out-diffusion occurs during annealing as discussed. The electron density in Ge is limited to a relatively low level, i.e., ~2×10¹⁹ cm⁻³ with even optimized annealing temperature and duration. One can foresee that the Ge direct band gap emission can be further improved if the dopant out-diffusion can be prevented or decreased. In this work, we employed different diffusion blocking layers on top of Ge to prevent dopant loss and thus obtain higher electron density. SiO₂, Si₃N₄, and α-Si with the thickness of 100 nm were deposited as capping layers on top of Ge. Secondary ion mass spectroscopy (SIMS) measurements have been carried out to determine the P concentration profile in the Ge. Figure 6 shows the SIMS results for the epi-Ge on Si annealed with various capping layers. For comparison, P profile of the as-implanted sample and the sample annealed without capping layer are also included. All the samples were annealed at 700 °C for 300 seconds.

Fig. 6 Comparison among P profiles in P-implanted Ge annealed with various capping layers, i.e., SiO₂, α-Si, and S_i3N₄. P profiles of the as-implanted sample and the sample annealed without capping layer are also shown in the figure. All the samples were annealed with the same condition, i.e., at 700 oC for 300 seconds.

As discussed above, the sample annealed without a capping layer would only have a total activated ions of ~23% out of the implanted ions. From SIMS results, we can calculate that the P concentration retained in Ge after annealing is ~36%, thus the unactivated P in Ge is ~13% out of the total implanted ions. Similarly, we also estimated the fractions of out-diffused ions, unactivated ions, and activated ions for the samples annealed with different capping layers (i.e., SiO₂, α-Si, and Si₃N₄). The result is summarized in Table 1 . The sheet resistance was also examined to verify the effect of capping layers on the electrical properties of annealed P-implanted Ge. As shown in Fig. 7 , the Si₃N₄ capped samples shows the lowest sheet resistance after annealing, indicating that the Si₄N₃ is the most efficient in blocking dopant out-diffusion among these three materials. It is consistent with the result shown in Table 1. This is also in accordance with the findings presented in Ref [14

]. which studied the phosphorous loss and diffusion by comparing Si₃N₄ and SiO₂ films as passivation layers on Ge.

Table 1 Fractions of Out-Diffused Ions, Unactivated Ions, and Activated Ions for the Samples Annealed with Various Capping Layers (i.e., SiO₂, α-Si, and Si₃N₄)

	Out-diffused (%)	Retained in Ge
	Out-diffused (%)	Unactivated (%)			Activated (%)
No capping layer	64%		13%	23%
Capped with SiO₂	36%		36%	28%
Capped with α-Si	39%		34%	27%
Capped with Si₃N₄	19%		39%	42%

Fig. 7 (a) Sheet resistance of P⁺-implanted Ge epi-films as a function of annealing temperature for the samples capped with various capping layers. (b) Sheet resistance of P⁺-implanted Ge epi-films as a function of annealing time for the samples capped with various capping layers.

The PL spectra were measured for all the samples with different capping layers. The result shows a similar evolution with annealing temperature and time for each kind of sample, i.e., the strongest PL is observed for the sample annealed at 700 °C for 300 seconds for different capping layers. Figure 8(a) shows the PL spectra of the samples capped with different materials annealed at 700 °C for 300 second which was considered as the optimized annealing condition as discussed above. It should be pointed out that the PL spectra shown in Fig. 8(a) are taken directly from the sample with capping layers. The capping layers can optically influence both the reflectance the incident optical power and light extraction efficiency during PL measurement, so the PL spectra obtained directly from the samples do not reflect the real PL properties of Ge underlying the capping layer. Therefore, the PL spectra need to be corrected for the comparison among the samples coated with different capping layers. Figure 8(b) presents the corrected PL spectra which remove the effect of capping layers on the reflectance of the incident optical power and extraction efficiency. As can be seen in Fig. 8(b), the Ge films with capping layers show much stronger PL intensities than that of the sample without any capping layer. The intensitites enhancement would be explained by the decrease in dopant loss and increase of dopant activation efficiency due to the existence of capping layers during annealing, as shown in Table 1. The sample with Si₃N₄ capping layer presents the strongest PL peak intensity, i.e., ~3 times of the value of the un-capped sample. The enhancement of PL emission by passivating the Ge with the capping layer is associated with the increased electron density in Ge due to the blocking effect of capping layer on dopant out-diffusion. Other contributing factors may include the Ge crystalline quality improvement due to the presence of capping layer during annealing, which were reported in Ref [13

C. H. Poon, L. S. Tan, B. J. Cho, and A. Y. Du, “Dopant loss mechanism in n⁺/p germanium junctions during rapid thermal annealing,” J. Electrochem. Soc. 152(12), G895–G899 (2005). [CrossRef]

,14

Fig. 8 Directly measured PL spectra (a) and PL spectra after correction to the reflectance of pumping laser and light-out extraction efficiency of the P⁺-implanted Ge samples annealed with various capping layers. All samples were annealed at 700 °C for 300 seconds.

There is another observation shown in Fig. 8 that may attract one’s interest. It can be seen that not only the PL intensities are enhanced by passivating films on Ge during annealing but also the PL peak position is found to be shifted depending on the capping layer materials. The insets of Fig. 6(b) shows the PL peak wavelengths for the samples with different capping layers. The PL peaks of the samples with SiO₂ and without any capping layer are the same, i.e., 1550 nm, indicating these two samples are almost strain free. However, the α-Si capped samples shows PL peak at 1575 nm, while the PL of Si₃N₄ capped Ge is peaked at 1535 nm. The shifting of PL peaks is related to change of Ge direct band gap. It is well known that the band diagram is influenced by both the type and the level of strain in Ge [4

–6

,22

M. El Jurdi, M. de Kersauson, D. David, X. Checoury, G. Beaudoin, R. Jakomin, I. Sagnes, S. Sauvage, G. Fishman, and P. Boucaud, “Stimulated emission in single tensile-strained Ge photonic wire,” in Proceedings of 8th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, England, 2011), pp. ThC8.

]. As mentioned in the experiment section, the capping layers have different types and values of stress, and thus they not only act as dopant out-diffusion blocking layers but also as film stressors for Ge. For example, the Si₃N₄ capping layer has a tensile stress of 310-MPa leading to a compressive strain in Ge, while α-Si capping layer has a compressive stress of 360-MPa leading to a tensile strain in Ge. The additional strain induced by capping layers is confirmed with Raman measurement. Figure 9 shows the high resolution Raman spectra of the samples annealed with various capping layers. Using the same method mentioned earlier in this study, we estimate that the Ge has the tensile stress of 0.12%, 0.09%, 0.28%, and compressive stress of 0.14%, respectively, for the samples annealed without capping layers, with SiO₂, with α-Si, and with Si₃N₄ as capping layers. This explains the observation that PL peak position blue- and red- shift with Si₃N₄ and α-Si capping layers, respectively. Considering the fact the SiO₂ film has very low stress, i.e., 21-MPa compressive stress, the tensile strain in Ge changes very little in comparison with the un-capped sample, thus the PL peak of the sample annealed with SiO₂ capping layer shows no obvious shift compared with the sample without capping layers.

Fig. 9 High resolution Raman spectra of P-implanted Ge epi-film with various capping layers. All samples were annealed at 700 °C for 300 seconds.

3.4. Effect of implantation dose on the photoluminescence

As well known, the effective dopant concentration is the most important factor to influence the emission efficiency of n-type Ge film. With ion implantation technique, the dopant density is controlled by the implantation dose. However, the effective n-type dopant concentration is limited by the solid solubility of n-type dopants in Ge. As such, a study on the relationship between the PL intensity and the implantation dose is important. In this experiment, two-energy implantation is carried out at fixed energies of 30 and 80 keV, while the total dose varying from 2 × 10¹⁴ to 4 × 10¹⁵ cm⁻² (note that the total dose was implanted twice with each sub-dose being half amount of the total dose). In order to have the most efficient PL emissions, Si₃N₄-capped P-implanted Ge annealed at 700 °C for 300 seconds is investigated in this study as this condition shows strongest PL as discussed above. Figure 10 shows the PL spectra of P-implanted Ge with varying implanted dose. As can be seen in the figure, the strongest PL is found from the samples implanted with the dose of 4 × 10¹⁴ cm⁻² while a continuous drop can be observed with increasing implantation dose. This could be because that more severe lattice damage happens when higher dose of phosphorous is implanted. Based on the PL result shown in the figure, an optimized implantation dose should be in the range of 2 × 10¹⁴ to 1 × 10¹⁵ cm⁻² and close to 4 × 10¹⁴ cm⁻². The highest effective phosphorous concentration activated by thermal annealing in Ge is 5 × 10¹⁹ cm⁻³ as reported so far [12

C. O. Chui, L. Kulig, J. Moran, W. Tsai, and K. C. Saraswat, “Germanium n-type shallow junction activation dependences,” Appl. Phys. Lett. 87(9), 091909 (2005). [CrossRef]

]. A quick calculation assuming no dopant loss and flat dopant profile after annealing with the implantation dose of 1 × 10¹⁵ cm⁻² would induce an effective average n-type concentrating of 5 × 10¹⁹ cm⁻³ in 200-nm-thick Ge. As can be seen in the Fig. 10, however, this condition does not lead to the strongest PL among all the samples implanted with various doses. On the other hand, the sample implanted with less does (i.e., 4 × 10¹⁴ cm⁻²) shows the highest PL peak intensity. With a similar calculation, an implantation dose of 4 × 10¹⁴ cm⁻² can induce an average effective n-type concentration of 2 × 10¹⁹ cm⁻³. Obviously, the effective P concentration which induces the strongest PL in this work is much less than what have been reported in Ref [12

C. O. Chui, L. Kulig, J. Moran, W. Tsai, and K. C. Saraswat, “Germanium n-type shallow junction activation dependences,” Appl. Phys. Lett. 87(9), 091909 (2005). [CrossRef]

]. This could be explained in the way that Ge crystal quality is degraded dramatically due to the implant-induced damage while the implantation dose is higher than 4 × 10¹⁴ cm⁻². This finding implicates that there has to be a trade-off between implantation dose and Ge crystal quality to have best luminescence performance of P-implanted Ge.

Fig. 10 PL spectra of the samples with various implantation doses.

3.5. Discussion

Based on all findings discussed above, process guidelines can be proposed for achieving best PL performance from P-implanted Ge epitaxial Ge thin film. First, the annealing condition should be optimized to deal with trade-off among dopant activation and dopant loss. In this work, it shows that annealing at 700 °C for 300 seconds induces strongest PL from P-implanted Ge film. Secondly, suitable out-diffusion block layers should be applied to prevent dopant loss during annealing. We have found that Si₃N₄ capping is most efficient in preventing dopant loss during annealing than SiO₂ and α-Si film. Moreover, it has been found that capping film not only act as diffusion block layer but also stress the Ge film to have additional strain which can impact the PL properties. In terms of additional strain in Ge, α-Si provides the tensile strain which is desired to enhance the luminescence efficiency of Ge. However, among the material we have studied, no capping layers can provide both efficient blocking effect of dopant out-diffusion and additional tensile strain in Ge. Before that kind material can be developed, we propose a process flow described as follows to optimize the PL intensity. The P-implanted Ge is first capped with Si₃N₄ film and then annealed at 700 °C for 300 seconds. After annealing, Si₃N₄ is etched away using reactive ion dry etching followed by α-Si deposition to make Ge have extra tensile strain. Figure 11 shows the PL spectra from the sample processed with above procedures. The PL spectra of the sample annealed without capping layer and annealed with Si₃N₄ capping layer are also included for comparison. Spectra shown in Fig. 11 are correlated to the reflectance of the pumping laser and extraction efficiency of emission generated by the pumping. It can be found that the proposed process make Ge emit most efficiently, i.e., ~4 times of that the Ge annealed with optimized temperature and time without capping layer. At the same time, the PL peak shifts to 1575 nm due to the additional tensile generated by α-Si stressor.

Fig. 11 PL spectra of the samples annealed without any capping layer, annealed with Si₃N₄ capping layer, and after Si₃N₄ etching and α-Si deposition.

4. Summary

In conclusion, we have demonstrated a systematic study on the dependences of PL from P-implanted epitaxial Ge on Si. Rapid thermal annealing has been performed for various annealing temperatures and times. The evolution of photoluminescence of the P-implanted Ge with thermal annealing has been investigated. It has been studied and correlated to sheet resistance evolution which indicates the electron density in the film. The strongest PL intensity is achieved with the optimized annealing condition, i.e., 700 °C for 300 seconds. Furthermore, significant enhancement of PL intensity has been observed by depositing a film as dopant out-diffusion blocking layer during annealing. Various blocking layers have been studied including PECVD SiO₂, PECVD Si₃N₄, and LPCVD α-Si. Among these, Si₃N₄ is found to be the most efficient film in blocking dopant out-diffusion, and thus leads to the strongest PL intensity. Another finding is that the capping film not only reduces the dopant out-diffusion but also act as a stressor making Ge have extra strain which can blue- or red- shift the PL spectrum. Moreover, a trade-off is found between the implantation does and PL intensity. Based on the findings above, we proposed optimized process flow maximizing the PL intensity from P-implanted epitaxial Ge film. The result shows that the optimized process increases the PL intensity by ~4 times as compared with that of the sample annealed without capping layer.

References and links

1.	L. Chen, P. Dong, and M. Lipson, “High performance germanium photodetectors integrated on submicron silicon waveguides by low temperature wafer bonding,” Opt. Express 16(15), 11513–11518 (2008). [CrossRef] [PubMed]
2.	K.-W. Ang, M.-B. Yu, S.-Y. Zhu, K.-T. Chua, G.-Q. Lo, and D.-L. Kwong, “Novel NiGe MSM photodetector featuring asymmetrical Schottky barriers using sulfur Co-implantation and segregation,” IEEE Electron Device Lett. 29(7), 708–711 (2008). [CrossRef]
3.	K.-W. Ang, M.-B. Yu, G.-Q. Lo, and D.-L. Kwong, “Low voltage and high responsivity germanium bipolar phototransistor for optical detections in the near-infrared regime,” IEEE Electron Device Lett. 29(10), 1124–1127 (2008). [CrossRef]
4.	J. Liu, X. Sun, D. Pan, X. Wang, L. C. Kimerling, T. L. Koch, and J. Michel, “Tensile-strained, n-type Ge as a gain medium for monolithic laser integration on Si,” Opt. Express 15(18), 11272–11277 (2007). [CrossRef] [PubMed]
5.	X. Sun, J. Liu, L. C. Kimerling, J. Michel, and T. L. Koch, “Direct gap photoluminescence of n-type tensile-strained Ge-on-Si,” Appl. Phys. Lett. 95(1), 011911 (2009). [CrossRef]
6.	J. Liu, X. Sun, L. C. Kimerling, and J. Michel, “Towards a Ge-based laser for CMOS applications,” in Proceedings of 5^th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, Italy, 2008), pp. 16–18.
7.	M. El Kurdi, T. Kociniewski, T.-P. Ngo, J. Boulmer, D. Débarre, P. Boucaud, J. F. Damlencourt, O. Kermarrec, and D. Bensahel, “Enhanced photoluminescence of heavily n-doped germanium,” Appl. Phys. Lett. 94(19), 191107 (2009). [CrossRef]
8.	T.-H. Cheng, K.-L. Peng, C.-Y. Ko, C.-Y. Chen, H.-S. Lan, Y.-R. Wu, C. W. Liu, and H.-H. Tseng, “Strain-enhanced photoluminescence from Ge direct transition,” Appl. Phys. Lett. 96(21), 211108 (2010). [CrossRef]
9.	X. Sun, J. Liu, L. C. Kimerling, and J. Michel, “Room-temperature direct bandgap electroluminesence from Ge-on-Si light-emitting diodes,” Opt. Lett. 34(8), 1198–1200 (2009). [CrossRef] [PubMed]
10.	S.-L. Cheng, J. Lu, G. Shambat, H.-Y. Yu, K. Saraswat, J. Vuckovic, and Y. Nishi, “Room temperature 1.6 microm electroluminescence from Ge light emitting diode on Si substrate,” Opt. Express 17(12), 10019–10024 (2009). [CrossRef] [PubMed]
11.	C. O. Chui, K. Gopalakrishnan, P. B. Griffin, J. D. Plummer, and K. C. Saraswat, “Activation and diffusion studies of ion-implanted p and n dopants in germanium,” Appl. Phys. Lett. 83(16), 3275–3277 (2003). [CrossRef]
12.	C. O. Chui, L. Kulig, J. Moran, W. Tsai, and K. C. Saraswat, “Germanium n-type shallow junction activation dependences,” Appl. Phys. Lett. 87(9), 091909 (2005). [CrossRef]
13.	C. H. Poon, L. S. Tan, B. J. Cho, and A. Y. Du, “Dopant loss mechanism in n⁺/p germanium junctions during rapid thermal annealing,” J. Electrochem. Soc. 152(12), G895–G899 (2005). [CrossRef]
14.	A. Chroneos, D. Skarlatos, C. Tsamis, A. Christofi, D. S. McPhail, and R. Hung, “Implantation and diffusion of phosphorous in germanium,” Mater. Sci. Semicond. Process. 9(4-5), 640–643 (2006). [CrossRef]
15.	T. H. Loh, H. S. Nguyen, C. H. Tung, A. D. Trigg, G. Q. Lo, N. Balasubramanian, D. L. Kwong, and S. Tripathy, “Ultrathin low temperature SiGe buffer for the growth of high quality Ge epilayer on Si(100) by ultrahigh vacuum chemical vapor deposition,” Appl. Phys. Lett. 90(9), 092108 (2007). [CrossRef]
16.	G. Grzybowski, R. Roucka, J. Mathews, L. Jiang, R. T. Beeler, J. Kouvetakis, and J. Menendez, “Direct versus indirect optical recombination in Ge films grown on Si substrate,” Phys. Rev. B 84(20), 205307 (2011). [CrossRef]
17.	D. D. Cannon, J. Liu, Y. Ishikawa, K. Wada, D. T. Danielson, S. Jongthammanurak, J. Michel, and L. C. Kimerling, “Tensile strained epitaxial Ge films on Si(100) substrates with potential application in L-band telecommunications,” Appl. Phys. Lett. 84(6), 906–908 (2004). [CrossRef]
18.	F. Cerdeira, C. Buchenauer, F. Pollak, and M. Cardona, “Stress-induced shifts of first-order Raman frequencies of diamond- and Zinc-blende-type semiconductors,” Phys. Rev. B 5(2), 580–593 (1972). [CrossRef]
19.	S. S. Mitra and N. E. Massa, Handbook of Semiconductors, T. S. Moss ed. (North-Holland, 1986) Vol. 1, p. 96.
20.	J. Joo, S. Kim, I. G. Kim, K. S. Jang, and G. Kim, “High-sensitivity 10 Gbps Ge-on-Si photoreceiver operating at lambda approximately 1.55 microm,” Opt. Express 18(16), 16474–16479 (2010). [CrossRef] [PubMed]
21.	S. M. Sze and J. C. Irvin, “Resistivity, mobility and impurity levels in GaAs, Ge, and Si at 300 °K,” Solid-State Electron. 11(6), 599–602 (1968). [CrossRef]
22.	M. El Jurdi, M. de Kersauson, D. David, X. Checoury, G. Beaudoin, R. Jakomin, I. Sagnes, S. Sauvage, G. Fishman, and P. Boucaud, “Stimulated emission in single tensile-strained Ge photonic wire,” in Proceedings of 8th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, England, 2011), pp. ThC8.

OCIS Codes
(130.3130) Integrated optics : Integrated optics materials
(140.3380) Lasers and laser optics : Laser materials

ToC Category:
Integrated Optics

History
Original Manuscript: November 14, 2011
Revised Manuscript: February 1, 2012
Manuscript Accepted: February 1, 2012
Published: March 26, 2012

Citation
L. Ding, Andy Eu-Jin Lim, Jason Tsung-Yang Liow, M. B. Yu, and G.-Q. Lo, "Dependences of photoluminescence from P-implanted epitaxial Ge," Opt. Express 20, 8228-8239 (2012)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-20-8-8228

Sort: Author | Year | Journal | Reset

References

L. Chen, P. Dong, and M. Lipson, “High performance germanium photodetectors integrated on submicron silicon waveguides by low temperature wafer bonding,” Opt. Express16(15), 11513–11518 (2008). [CrossRef] [PubMed]
K.-W. Ang, M.-B. Yu, S.-Y. Zhu, K.-T. Chua, G.-Q. Lo, and D.-L. Kwong, “Novel NiGe MSM photodetector featuring asymmetrical Schottky barriers using sulfur Co-implantation and segregation,” IEEE Electron Device Lett.29(7), 708–711 (2008). [CrossRef]
K.-W. Ang, M.-B. Yu, G.-Q. Lo, and D.-L. Kwong, “Low voltage and high responsivity germanium bipolar phototransistor for optical detections in the near-infrared regime,” IEEE Electron Device Lett.29(10), 1124–1127 (2008). [CrossRef]
J. Liu, X. Sun, D. Pan, X. Wang, L. C. Kimerling, T. L. Koch, and J. Michel, “Tensile-strained, n-type Ge as a gain medium for monolithic laser integration on Si,” Opt. Express15(18), 11272–11277 (2007). [CrossRef] [PubMed]
X. Sun, J. Liu, L. C. Kimerling, J. Michel, and T. L. Koch, “Direct gap photoluminescence of n-type tensile-strained Ge-on-Si,” Appl. Phys. Lett.95(1), 011911 (2009). [CrossRef]
J. Liu, X. Sun, L. C. Kimerling, and J. Michel, “Towards a Ge-based laser for CMOS applications,” in Proceedings of 5th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, Italy, 2008), pp. 16–18.
M. El Kurdi, T. Kociniewski, T.-P. Ngo, J. Boulmer, D. Débarre, P. Boucaud, J. F. Damlencourt, O. Kermarrec, and D. Bensahel, “Enhanced photoluminescence of heavily n-doped germanium,” Appl. Phys. Lett.94(19), 191107 (2009). [CrossRef]
T.-H. Cheng, K.-L. Peng, C.-Y. Ko, C.-Y. Chen, H.-S. Lan, Y.-R. Wu, C. W. Liu, and H.-H. Tseng, “Strain-enhanced photoluminescence from Ge direct transition,” Appl. Phys. Lett.96(21), 211108 (2010). [CrossRef]
X. Sun, J. Liu, L. C. Kimerling, and J. Michel, “Room-temperature direct bandgap electroluminesence from Ge-on-Si light-emitting diodes,” Opt. Lett.34(8), 1198–1200 (2009). [CrossRef] [PubMed]
S.-L. Cheng, J. Lu, G. Shambat, H.-Y. Yu, K. Saraswat, J. Vuckovic, and Y. Nishi, “Room temperature 1.6 microm electroluminescence from Ge light emitting diode on Si substrate,” Opt. Express17(12), 10019–10024 (2009). [CrossRef] [PubMed]
C. O. Chui, K. Gopalakrishnan, P. B. Griffin, J. D. Plummer, and K. C. Saraswat, “Activation and diffusion studies of ion-implanted p and n dopants in germanium,” Appl. Phys. Lett.83(16), 3275–3277 (2003). [CrossRef]
C. O. Chui, L. Kulig, J. Moran, W. Tsai, and K. C. Saraswat, “Germanium n-type shallow junction activation dependences,” Appl. Phys. Lett.87(9), 091909 (2005). [CrossRef]
C. H. Poon, L. S. Tan, B. J. Cho, and A. Y. Du, “Dopant loss mechanism in n+/p germanium junctions during rapid thermal annealing,” J. Electrochem. Soc.152(12), G895–G899 (2005). [CrossRef]
A. Chroneos, D. Skarlatos, C. Tsamis, A. Christofi, D. S. McPhail, and R. Hung, “Implantation and diffusion of phosphorous in germanium,” Mater. Sci. Semicond. Process.9(4-5), 640–643 (2006). [CrossRef]
T. H. Loh, H. S. Nguyen, C. H. Tung, A. D. Trigg, G. Q. Lo, N. Balasubramanian, D. L. Kwong, and S. Tripathy, “Ultrathin low temperature SiGe buffer for the growth of high quality Ge epilayer on Si(100) by ultrahigh vacuum chemical vapor deposition,” Appl. Phys. Lett.90(9), 092108 (2007). [CrossRef]
G. Grzybowski, R. Roucka, J. Mathews, L. Jiang, R. T. Beeler, J. Kouvetakis, and J. Menendez, “Direct versus indirect optical recombination in Ge films grown on Si substrate,” Phys. Rev. B84(20), 205307 (2011). [CrossRef]
D. D. Cannon, J. Liu, Y. Ishikawa, K. Wada, D. T. Danielson, S. Jongthammanurak, J. Michel, and L. C. Kimerling, “Tensile strained epitaxial Ge films on Si(100) substrates with potential application in L-band telecommunications,” Appl. Phys. Lett.84(6), 906–908 (2004). [CrossRef]
F. Cerdeira, C. Buchenauer, F. Pollak, and M. Cardona, “Stress-induced shifts of first-order Raman frequencies of diamond- and Zinc-blende-type semiconductors,” Phys. Rev. B5(2), 580–593 (1972). [CrossRef]
S. S. Mitra and N. E. Massa, Handbook of Semiconductors, T. S. Moss ed. (North-Holland, 1986) Vol. 1, p. 96.
J. Joo, S. Kim, I. G. Kim, K. S. Jang, and G. Kim, “High-sensitivity 10 Gbps Ge-on-Si photoreceiver operating at lambda approximately 1.55 microm,” Opt. Express18(16), 16474–16479 (2010). [CrossRef] [PubMed]
S. M. Sze and J. C. Irvin, “Resistivity, mobility and impurity levels in GaAs, Ge, and Si at 300 °K,” Solid-State Electron.11(6), 599–602 (1968). [CrossRef]
M. El Jurdi, M. de Kersauson, D. David, X. Checoury, G. Beaudoin, R. Jakomin, I. Sagnes, S. Sauvage, G. Fishman, and P. Boucaud, “Stimulated emission in single tensile-strained Ge photonic wire,” in Proceedings of 8th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, England, 2011), pp. ThC8.

Appl. Phys. Lett.

X. Sun, J. Liu, L. C. Kimerling, J. Michel, and T. L. Koch, “Direct gap photoluminescence of n-type tensile-strained Ge-on-Si,” Appl. Phys. Lett.95(1), 011911 (2009). [CrossRef]
M. El Kurdi, T. Kociniewski, T.-P. Ngo, J. Boulmer, D. Débarre, P. Boucaud, J. F. Damlencourt, O. Kermarrec, and D. Bensahel, “Enhanced photoluminescence of heavily n-doped germanium,” Appl. Phys. Lett.94(19), 191107 (2009). [CrossRef]
T.-H. Cheng, K.-L. Peng, C.-Y. Ko, C.-Y. Chen, H.-S. Lan, Y.-R. Wu, C. W. Liu, and H.-H. Tseng, “Strain-enhanced photoluminescence from Ge direct transition,” Appl. Phys. Lett.96(21), 211108 (2010). [CrossRef]
C. O. Chui, K. Gopalakrishnan, P. B. Griffin, J. D. Plummer, and K. C. Saraswat, “Activation and diffusion studies of ion-implanted p and n dopants in germanium,” Appl. Phys. Lett.83(16), 3275–3277 (2003). [CrossRef]
C. O. Chui, L. Kulig, J. Moran, W. Tsai, and K. C. Saraswat, “Germanium n-type shallow junction activation dependences,” Appl. Phys. Lett.87(9), 091909 (2005). [CrossRef]
T. H. Loh, H. S. Nguyen, C. H. Tung, A. D. Trigg, G. Q. Lo, N. Balasubramanian, D. L. Kwong, and S. Tripathy, “Ultrathin low temperature SiGe buffer for the growth of high quality Ge epilayer on Si(100) by ultrahigh vacuum chemical vapor deposition,” Appl. Phys. Lett.90(9), 092108 (2007). [CrossRef]
D. D. Cannon, J. Liu, Y. Ishikawa, K. Wada, D. T. Danielson, S. Jongthammanurak, J. Michel, and L. C. Kimerling, “Tensile strained epitaxial Ge films on Si(100) substrates with potential application in L-band telecommunications,” Appl. Phys. Lett.84(6), 906–908 (2004). [CrossRef]

IEEE Electron Device Lett.

K.-W. Ang, M.-B. Yu, S.-Y. Zhu, K.-T. Chua, G.-Q. Lo, and D.-L. Kwong, “Novel NiGe MSM photodetector featuring asymmetrical Schottky barriers using sulfur Co-implantation and segregation,” IEEE Electron Device Lett.29(7), 708–711 (2008). [CrossRef]
K.-W. Ang, M.-B. Yu, G.-Q. Lo, and D.-L. Kwong, “Low voltage and high responsivity germanium bipolar phototransistor for optical detections in the near-infrared regime,” IEEE Electron Device Lett.29(10), 1124–1127 (2008). [CrossRef]

J. Electrochem. Soc.

C. H. Poon, L. S. Tan, B. J. Cho, and A. Y. Du, “Dopant loss mechanism in n+/p germanium junctions during rapid thermal annealing,” J. Electrochem. Soc.152(12), G895–G899 (2005). [CrossRef]

Mater. Sci. Semicond. Process.

A. Chroneos, D. Skarlatos, C. Tsamis, A. Christofi, D. S. McPhail, and R. Hung, “Implantation and diffusion of phosphorous in germanium,” Mater. Sci. Semicond. Process.9(4-5), 640–643 (2006). [CrossRef]

Opt. Express

Opt. Lett.

X. Sun, J. Liu, L. C. Kimerling, and J. Michel, “Room-temperature direct bandgap electroluminesence from Ge-on-Si light-emitting diodes,” Opt. Lett.34(8), 1198–1200 (2009). [CrossRef] [PubMed]

Phys. Rev. B

F. Cerdeira, C. Buchenauer, F. Pollak, and M. Cardona, “Stress-induced shifts of first-order Raman frequencies of diamond- and Zinc-blende-type semiconductors,” Phys. Rev. B5(2), 580–593 (1972). [CrossRef]
G. Grzybowski, R. Roucka, J. Mathews, L. Jiang, R. T. Beeler, J. Kouvetakis, and J. Menendez, “Direct versus indirect optical recombination in Ge films grown on Si substrate,” Phys. Rev. B84(20), 205307 (2011). [CrossRef]

Solid-State Electron.

S. M. Sze and J. C. Irvin, “Resistivity, mobility and impurity levels in GaAs, Ge, and Si at 300 °K,” Solid-State Electron.11(6), 599–602 (1968). [CrossRef]

Other

M. El Jurdi, M. de Kersauson, D. David, X. Checoury, G. Beaudoin, R. Jakomin, I. Sagnes, S. Sauvage, G. Fishman, and P. Boucaud, “Stimulated emission in single tensile-strained Ge photonic wire,” in Proceedings of 8th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, England, 2011), pp. ThC8.
S. S. Mitra and N. E. Massa, Handbook of Semiconductors, T. S. Moss ed. (North-Holland, 1986) Vol. 1, p. 96.
J. Liu, X. Sun, L. C. Kimerling, and J. Michel, “Towards a Ge-based laser for CMOS applications,” in Proceedings of 5th IEEE International. Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, Italy, 2008), pp. 16–18.

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.

Click here to see a list of articles that cite this paper